Mobile-based expert system for human diet planning using optimum neighbor

Abstract

This research proposes an expert system method to recommend the quantity of every ingredients food for a normal human or specific diet patient. Our proposed method initial state was 100 pairs of generated random value. Afterward, the pair of value which contains minimum error rate was chosen. Our proposed method uses the generated optimum neighbor as the recommendation solution. Our proposed method was implemented as an android application, named SlimLine. Based on the experiment, SlimLine able to compose the food ingredients quantity with the macronutrient needs in the range about 25% above or below nutrition needs.